This proposal aims to use fMRI as an index of human cortical activity when viewing colored objects. Color tuning will be measured in different regions of visual cortex. The neural basis of two main color phenomena will studied. The first question is how the visual system compensates for changes in the ambient illumination and computes a relatively constant color appearance for surfaces. The second study address the problem of why colored patterns at high temporal or spatial frequency appear achromatic.